A MULTIPLE-BASE-LINE STEREO

This paper presents a stereo matching method that uses multiple stereo pairs with various baselines to obtain precise distance estimates without suffering from ambiguity. In stereo processing, a short baseline means that the estimated distance will be less precise due to narrow triangulation. For more precise distance estimation, a longer baseline is desired. With a longer baseline, however, a larger disparity range must be searched to find a match. As a result, matching is more difficult, and there is a greater possibility of a false match. Therefore, there is a tradeoff between precision and accuracy in matching. The stereo matching method presented in this paper uses multiple stereo pairs with different baselines generated by a lateral displacement of a camera. Matching is performed simply by computing the sum of squared-difference (SSD) values. The SSD functions for individual stereo pairs are represented with respect to the inverse distance (rather than the disparity, as is usually done) and are then simply added to produce the sum of SSD's. This resulting function is called the SSSD-in-inverse-distance. We show that the SSSD-in-inverse-distance function exhibits a unique and clear minimum at the correct matching position, even when the underlying intensity patterns of the scene include ambiguities or repetitive patterns. An advantage of this method is that we can eliminate false matches and increase precision without any search or sequential filtering. This paper first defines a stereo algorithm based on the SSSD-in-inverse-distance and presents a mathematical analysis to show how the algorithm can remove ambiguity and increase precision. Then, a few experimental results with real stereo images are presented to demonstrate the effectiveness of the algorithm.